This proposal involves five new, long-term goals: 1) we will study the biochemical mechanism of capillary basement membrane thickening, resembling that of diabetes, which we have found occurs in the retinal and cerebral microvessels of rats with galactosemia induced by diet, and is prevented by an aldose reductase inhibitor. Quantitative and qualitative aspects of basement membrane collagen, proteoglycan (PG), and laminin synthesis will be evaluated following (3H)-proline and (35S)-sulfate administration, using biochemical analysis and enzyme-linked immunosorbent assay. 2) We will compare the biosynthesis of PGs of the interphotoreceptor matrix (IPM) in spontaneously hypertensive (SHR) rats, which develop a photoreceptor dystrophy, and in genetically closely related Wistar-Kyoto (WKy) animals, which do not. We have found that the dystrophy in SHR animals is greatly retarded with a 30% galactose diet which, we propose, may alter PG aynthesis in the IPM. (35S)-sulfate and (3H)-glucosamine will be administered for autoradiographic and biochemical analyses of retinal PGs in SHR and WKy animals on high galactose and normal diets. 3) We will evaluate the proposed contractile function of pericytes in the retinal and cerebral circulation of normal, diabetic, and galactosemic rats by comparing them in the normal state and following administration of agents that induce vascular smooth muscle contraction. Electron microscope morphometric methods will be used to assess contractility of pericytes and/or endothelial cells. 4) We will evaluate the integrity of the blood-brain and blood-retinal barriers in diabetic, galactosemic, and normal rats following production of acute hypertension pharmacologically. This will be done by evaluating gross leakage of Evans blue dye, and microscopic leakage of horseradish peroxidase. The studies in (3) and (4) will test the hypothesis that dysfunciton of pericytes may occur early in diabetes and galactosemia, well before pericyte loss occurs. The consequences of this dysfunction may be more prominent in the retinal microcirculation than in the cerebral because of the much larger proportion of pericytes to endothelial cells in the retina. 5) We will attempt to develop models of retinovitreal neovascularization in the rat with much higher prevalence than we observed in dystrophic SHR animals during the previous funding period. By sequential ophthalmoscopic observation and photography, and by (3H)-thymidine autoradiography, we will attempt to provide further evidence that these truly are new vessels, comparable to those of various human disease states.